Our long-term goal is the mechanistic dissection of neuropeptide signals in neural circuits in vivo. We have used a variety of genetic reagents for in vivo cell-specific manipulation of three Class B1 neuropeptides whose G protein-coupled receptors signal through cAMP in fly circadian/sleep control circuits: PDF (Pigment Dispersing Factor), DH31 (fly calcitonin homologue), and DH44 (fly corticotropin releasing factor (CRF) homologue). PDF is already known to be expressed in a subset of circadian clock neurons and to be important for regulating fly daily rhythms and sleep. Our preliminary studies now suggest distinct important roles for DH31 and DH44 in controlling daily patterns of sleep and activity. Flies adapt their bimodal crepuscular pattern of rest and activity to prevailing seasonal conditions: in the winter most activity is in the evening (to avoid the chill of night), while in the summer most activity is in the morning (to avoid the heat of day). Our preliminary studies suggest that increased autocrine activation of PDF receptors (PDFR) possessed by the PDF-secreting neurons themselves underlies this seasonal shift in the balance of activity from evening to morning. We will use various genetic tools to test the hypothesis that autocrine PDFR activation induces this plastic change in circadian network properties by modulating the daily pattern of PDF secretion itself. Recent studies implicate a particular subset of non-PDF-secreting clock cells as the required recipients of this PDF signal. Our preliminary studies indicate that many of the neurons in this subset express DH31, and that - like flies lacking PDF - mutant flies lacking DH31 exhibit severely blunted circadian morning activity. Furthermore, PDF-secreting clock neurons themselves possess DH31 receptors (DH31R), thus suggesting the hypothesis that a reciprocal feedback loop between PDF-secreting and DH31-secreting clock neurons drives morning activity. We will use various genetic tools to test this hypothesis. Our preliminary studies show that DH44-the fly homologue of CRF-is expressed at high levels in a subset of neurons in the pars intercerebralis, fly homologue of the hypothalamus. Furthermore, we have performed a preliminary behavioral genetic screen suggesting that - as for mammalian CRF - DH44 signaling to central brain neurons decreases total sleep amount and increases sleep fragmentation. Based on these preliminary findings, we will use a variety of genetic tools to identify the specific DH44 receptor-expressing neurons responsible for the regulation of sleep.